Using the information from here I can summarize my dataframe using timestamp information.
data <- read.csv("data.csv", header=T)
data$TIMESTAMP <- strptime(data$TIMESTAMP, "%m/%d/%Y %H:%M") # make unambigous
data$TIMESTAMP <- as.POSIXct(data$TIMESTAMP) # plyr/ddply does not seem to like POSIXlt
data$DAY <- as.factor(format(data$TIMESTAMP,'%d'))
data$MONTH <- as.factor(format(data$TIMESTAMP,'%m'))
ddply(data,.(MONTH,DAY),summarise, V1 = mean(P), V2 = max(WS)) # summarize by month by day
However, I only want to use certain hours of the day, for example between 13:00:00 and 16:00:00. I can subset the dataframe (e.g. create a new a dataframe that is a subset of the original df) if I do so before the as.POSIXct
line is executed. But, when I try the following sequence:
data.1.4 <- data[format(data$TIMESTAMP, "%H") >= 13 & format(data$TIMESTAMP, "%H") < 17,] # subset the data
data.1.4$TIMESTAMP <- as.POSIXct(data.1.4$TIMESTAMP)
data.1.4$DAY <- as.factor(format(data.1.4$TIMESTAMP,'%d'))
data.1.4$MONTH <- as.factor(format(data.1.4$TIMESTAMP,'%m'))
ddply(data.1.4,.(MONTH,DAY),summarise, V1 = mean(P), V2 = max(WS)) # summarize by month by day
I get the following error Error in eval(expr, envir, enclos) : object 'MONTH' not found
. The only difference I can see between data
and data.1.4
is that the row.names are 'visible' and non-sequential data.1.4
dataframe.
So, how might I subset or select entries from the dataframe to create a new df that I can then summarize?
# dput of subset of data
data <- structure(list(TIMESTAMP = structure(1:15, .Label = c("1/1/2012 11:00",
"1/1/2012 12:00", "1/1/2012 13:00", "1/1/2012 14:00", "1/1/2012 15:00",
"1/2/2012 11:00", "1/2/2012 12:00", "1/2/2012 13:00", "1/2/2012 14:00",
"1/2/2012 15:00", "4/7/2012 11:00", "4/7/2012 12:00", "4/7/2012 13:00",
"4/7/2012 14:00", "4/7/2012 15:00"), class = "factor"), P = c(992.4,
992.4, 992.4, 992.4, 992.4, 992.4, 992.4, 992.4, 992.4, 992.4,
239, 239, 239, 239, 239), WS = c(4.023, 3.576, 4.023, 6.259,
4.47, 3.576, 3.576, 2.682, 4.023, 3.576, 2.682, 3.129, 2.682,
2.235, 2.682), WD = c(212L, 200L, 215L, 213L, 204L, 304L, 276L,
273L, 307L, 270L, 54L, 24L, 304L, 320L, 321L), AT = c(16.11,
18.89, 20, 20, 19.44, 10.56, 11.11, 11.67, 12.22, 11.11, 17.22,
18.33, 19.44, 20.56, 21.11), FT = c(17.22, 22.22, 22.78, 22.78,
20, 11.11, 15.56, 17.22, 17.78, 15.56, 24.44, 25.56, 29.44, 30.56,
29.44), H = c(50L, 38L, 38L, 39L, 48L, 24L, 19L, 18L, 16L, 18L,
23L, 20L, 18L, 17L, 15L), B = c(1029L, 1027L, 1026L, 1024L, 1023L,
1026L, 1025L, 1024L, 1023L, 1023L, 1034L, 1033L, 1032L, 1031L,
1030L), FM = c(14.9, 14.4, 14, 13.7, 13.6, 13.1, 12.8, 12.3,
12, 11.7, 12.8, 12, 11.4, 10.9, 10.4), GD = c(204L, 220L, 227L,
222L, 216L, 338L, 311L, 326L, 310L, 273L, 62L, 13L, 312L, 272L,
281L), MG = c(8.047, 9.835, 10.28, 13.41, 11.18, 9.388, 8.941,
8.494, 9.835, 10.73, 6.706, 7.153, 8.047, 8.047, 7.6), SR = c(522L,
603L, 604L, 526L, 248L, 569L, 653L, 671L, 616L, 487L, 972L, 1053L,
1061L, 1002L, 865L), WS2 = c(2.235, 3.576, 4.47, 4.47, 5.364,
4.023, 2.682, 3.576, 3.576, 4.023, 3.129, 3.129, 3.576, 2.682,
3.129), WD2 = c(200L, 201L, 206L, 210L, 211L, 319L, 315L, 311L,
302L, 290L, 49L, 39L, 15L, 348L, 329L)), .Names = c("TIMESTAMP",
"P", "WS", "WD", "AT", "FT", "H", "B", "FM", "GD", "MG", "SR",
"WS2", "WD2"), class = "data.frame", row.names = c(NA, -15L))